Beta-carotene (BC), which is present in many fruits and vegetables, and is relatively non-toxic, has been identified as an important chemotherapeutic agent to reduce the risk of epithelial cancer, including cancer of the colon. The hypothesis of the present study is that BC, at appropriate tissue levels, affects the rate of colonic cell proliferation (CCP), as evaluated by specific biochemical and/or histological markers. Our preliminary data have shown that chronic BC intake significantly reduces ornithine decarboxylase (ODC) activity in colonic epithelium of patients with colon cancer. We have also observed a decrease in serum lipid peroxide levels in normal young subjects upon supplementation with 15 mg BC per day. In the present study we will determine whether BC supplementation affects biochemical and histological indices of CCP such as urinary polyamines, cell cycle kinetics (measured by monoclonal antibody Ki-67) and the sialic acid free dissacharide residue, beta-D-Gal(1-3)-D-GalNAc. A total of 100 subjects (45-75 years old) will participate in the first phase of this study. They will include 20 normal subjects and two groups of 40 patients, including one group with colon cancer and another group with adenomatous colonic polyps, a precancerous condition. Initially we will determine the nutritional status and BC nutriture of all the groups and will assay indices of CCP. Subsequently, patients will be randomized into BC (30 mg/day) or placebo treatment groups and will be followed monthly. At 3 months the subjects will be evaluated again for BC status and indices of CCP. Patients on BC in whom ODC activity will be normalized will be asked to stop taking the supplement and will be followed for 3 more months. The patients in whom colonic ODC activity has not responded to BC supplementation will be entered into the second phase of the study and receive a maximal dose of 180 mg BC per day for 3 months. By providing detailed information on BC serum and tissue levels along with biochemical indices of CCP in 3 different groups, these studies will enable us to assess the appropriateness of certain intermediate endpoints in chemopreventive trials and may help establish target tissue levels for a potential therapeutic agent (BC).